[TTS] Fix defaulting of use_log_energy

Signed-off-by: Ryan <rlangman@nvidia.com>
NVIDIA · May 15, 2023 · c3009e8 · c3009e8
1 parent af017be
commit c3009e8
Show file tree

Hide file tree

Showing 2 changed files with 3 additions and 2 deletions.
diff --git a/nemo/collections/tts/models/fastpitch.py b/nemo/collections/tts/models/fastpitch.py
@@ -153,6 +153,7 @@ def __init__(self, cfg: DictConfig, trainer: Trainer = None):
         speaker_emb_condition_prosody = cfg.get("speaker_emb_condition_prosody", False)
         speaker_emb_condition_decoder = cfg.get("speaker_emb_condition_decoder", False)
         speaker_emb_condition_aligner = cfg.get("speaker_emb_condition_aligner", False)
+        use_log_energy = cfg.get("use_log_energy", True)
         if n_speakers > 1 and "add" not in input_fft.cond_input.condition_types:
             input_fft.cond_input.condition_types.append("add")
         if speaker_emb_condition_prosody:
@@ -177,7 +178,7 @@ def __init__(self, cfg: DictConfig, trainer: Trainer = None):
             energy_embedding_kernel_size,
             cfg.n_mel_channels,
             cfg.max_token_duration,
-            cfg.use_log_energy,
+            use_log_energy,
         )
         self._input_types = self._output_types = None
         self.export_config = {

diff --git a/tests/collections/tts/parts/utils/test_tts_dataset_utils.py b/tests/collections/tts/parts/utils/test_tts_dataset_utils.py
@@ -18,11 +18,11 @@
 import pytest
 import torch
 
-from nemo.collections.tts.parts.utils.tts_dataset_utils import get_abs_rel_paths, get_audio_filepaths, normalize_volume
 from nemo.collections.tts.parts.utils.tts_dataset_utils import (
     filter_dataset_by_duration,
     get_abs_rel_paths,
     get_audio_filepaths,
+    normalize_volume,
     stack_tensors,
 )